From: Gary Howland <gary@systemics.com>
Date: Fri, 31 May 1996 06:33:35 +0800
To: cypherpunks@toad.com
Subject: Re: [crypto] crypto-protocols for trading card games
Message-ID: <199605301440.QAA29702@internal-mail.systemics.com>
MIME-Version: 1.0
Content-Type: text/plain



David F. Ogren wrote:

> Cards are not transferrable.  In order to make cards transferrable the
> game company must be able to invalidate cards which have been traded
> to others.  In other words if Alice wants to give a cards to Bob she
> must:
>
> 1. Contact the game company and tell them she wants to give the card
> to Bob. 2. The game company must issue a new card to Bob with a new
> serial number and with Bob's public key rather than Alice's. 3. The
> game company must invalidate Alice's old card.  Since there is no way
> that the game company can make sure all copies of the card have been
> destroyed it must create a "invalid serial number list" and have the
> players dial into that list everytime the game is played.

This is the double spending problem.

> Since step 3 is so costly to implement, I think it is unlikely that a
> cryptography-based trading card game will have tradable cards.

Given that untraceability of tradable cards is less of an issue than
with e-cash, why not have a central registry of the owners of the cards
(which would consist of the card details paired with the public key
of the owner)?  Admittedly this means the players must be on line,
but then we all know how difficult off line detection of double
spenders is.

For anyone who is _serious_ about starting work on such a game
system, I have a few pieces of Perl and Java code that would really
get you on your way - let me know if you are interested.


Gary
--
pub  1024/C001D00D 1996/01/22  Gary Howland <gary@systemics.com>
Key fingerprint =  0C FB 60 61 4D 3B 24 7D  1C 89 1D BE 1F EE 09 06